Context and Geometry Aware Voxel Transformer for Semantic Scene Completion

TL;DR

This paper introduces CGFormer, a novel voxel transformer that incorporates context and geometry awareness for improved semantic scene completion, achieving state-of-the-art results on benchmark datasets.

Contribution

It proposes a context and geometry aware voxel transformer with a context-aware query generator and 3D deformable cross-attention, enhancing semantic and geometric understanding in SSC.

Findings

Achieves state-of-the-art mIoU of 16.87 on SemanticKITTI.

Attains 20.05 mIoU on SSCBench-KITTI-360.

Outperforms methods using temporal images or larger backbones.

Abstract

Vision-based Semantic Scene Completion (SSC) has gained much attention due to its widespread applications in various 3D perception tasks. Existing sparse-to-dense approaches typically employ shared context-independent queries across various input images, which fails to capture distinctions among them as the focal regions of different inputs vary and may result in undirected feature aggregation of cross-attention. Additionally, the absence of depth information may lead to points projected onto the image plane sharing the same 2D position or similar sampling points in the feature map, resulting in depth ambiguity. In this paper, we present a novel context and geometry aware voxel transformer. It utilizes a context aware query generator to initialize context-dependent queries tailored to individual input images, effectively capturing their unique characteristics and aggregating information within the region of interest. Furthermore, it extend deformable cross-attention from 2D to 3D pixel space, enabling the differentiation of points with similar image coordinates based on their depth coordinates. Building upon this module, we introduce a neural network named CGFormer to achieve semantic scene completion. Simultaneously, CGFormer leverages multiple 3D representations (i.e., voxel and TPV) to boost the semantic and geometric representation abilities of the transformed 3D volume from both local and global perspectives. Experimental results demonstrate that CGFormer achieves state-of-the-art performance on the SemanticKITTI and SSCBench-KITTI-360 benchmarks, attaining a mIoU of 16.87 and 20.05, as well as an IoU of 45.99 and 48.07, respectively. Remarkably, CGFormer even outperforms approaches employing temporal images as inputs or much larger image backbone networks.